Mitochondrial dysfunction and oxidative stress play important roles in Parkinson's disease (PD). PGC-1alpha, a transcriptional coactivator, upregulates mitochondrial biogenesis and antioxidant defenses, and thus is an attractive target for neuroprotection in PD. Susceptibility to MPTP, a mitochondrial toxin, is increased in mice lacking PGC-1alpha, whereas overexpressing PGC-1alpha protects against an oxidative challenge in cell lines. Levels of expression of genes regulated by PGC-1alpha are low in substantia nigra (SN) neurons in early PD. Two important genetic causes of PD now have been linked to low PGC-1alpha. First, a recent study showed that loss of Parkin function leads to increased levels of novel protein called PARIS which transcriptionally inhibits expression of PGC-1alpha. And recently it was demonstrated that alpha-synuclein binds to the PGC-1alphha promoter and also suppresses its transcription. Together, these data strongly implicate a pathogenic role for low PGC-1alpha activity in PD, and raise the hope that correction of the PGC-1alpha deficit in PD will be neuroprotective. However, unexpectedly, we and others find that overexpressing PGC-1alpha at very high levels leads to reduced Bdnf and suppression of the dopaminergic phenotype. Our preliminary data suggest that this may result from suppression of Pitx3, a transcription factor that is critical for maintaining the dopaminergic phenotype and also for expression of Bdnf. Vulnerability to MPTP is increased by the very high levels of PGC-1alpha achieved using our AAV-PGC-1alpha vector. Thus, either low or very high levels of PGC-1alpha can be deleterious. Together, these data reveal that maintenance of PGC-1alpha activity levels within a therapeutic range is critical for the survival and function of dopaminergic neurons. We hypothesize that very high levels of PGC-1alpha lead to suppression of Pitx3, leading to loss of the dopaminergic phenotype and to enhanced vulnerability to MPTP due to loss of Bdnf. We further hypothesize that it will be possible to harness the neuroprotective potential of viral vector- mediated increases in PGC-1alpha activity while avoiding the potentially deleterious effects associated with very high levels of overexpression. We propose to test this by studying the impact in dopaminergic neurons on mitochondrial function, oxidative stress, the dopaminergic phenotype, and susceptibility to MPTP following more modest levels of upregulation of PGC-1alpha, or following co-expression of Pitx3 to prevent the deleterious effects of higher PGC-1alpha levels. The potential neuroprotective effects of Pitx3 on its own also will be studied. These experiments will test our hypothesis that suppression of Pitx3 mediates the PGC-1alpha-induced downregulation of Bdnf and of the dopaminergic phenotype. In addition, multiple gene therapy trials have been conducted in PD patients, and thus the proposed studies also will serve as initial tests of therapeutic strategies with translational potential.